Newly Published
Correspondence  |   January 2020
Measuring Childbirth Outcomes: Reply
Author Notes
  • Department of Anesthesiology and Perioperative Medicine, University of Rochester School of Medicine, Rochester, New York (L.G.G.).
  • Accepted for publication January 2, 2020.
    Accepted for publication January 2, 2020.×
Article Information
Correspondence   |   January 2020
Measuring Childbirth Outcomes: Reply
Anesthesiology Newly Published on January 29, 2020. doi:
Anesthesiology Newly Published on January 29, 2020. doi:
We thank Sitcov et al. for their comments on our article.1  In their letter, the authors state that although they “agree with the recommendation that administrative data be submitted by all hospitals” to a national obstetrical outcomes registry, they believe that “clinical data are superior” to administrative data, and that clinical data are feasible and are “worth the effort to make the United States a safer place to give birth.” In principle, we are in complete agreement that there is not only a strong business case, but also a strong ethical argument for collecting high-quality clinical data to use as the basis for improving population outcomes, not only in obstetrics but in all medicine. Quality measurement is at the heart of efforts to provide actionable feedback to hospitals, physicians, and other providers. Quality measurement is also the linchpin of efforts by the federal government and third-party payers to deliver more cost-efficient higher quality care. And quality measurement based on clinical data is likely more valid compared with measurements based on administrative data. But, the question is who will pay for the cost of manually abstracting records for 3.8 million births annually in the United States?2  We commend the authors for their efforts in developing an obstetrical registry using manually abstracted clinical data from 24 institutions.3  Such a registry can be an important tool for research and quality improvement, but to be an effective tool for parents throughout the country, clinical data are needed from most if not all hospitals. Doing so by manually abstracting records would be cost prohibitive. In fact, the federal government, which is the single largest health care payer in the United States, uses administrative and not manually abstracted clinical data to measure quality. We believe that the most cost-effective alternative to manually abstracted clinical data is to extract structured clinical data from the electronic medical record. To that end, the American College of Obstetricians and Gynecologists (Washington, D.C.) and the American Society of Anesthesiologists (Schaumburg, Illinois) have partnered to create the Maternal Quality Improvement Program4  (now renamed the Birth Registry) to serve as a national platform for measuring and improving childbirth outcomes. The data elements in the data dictionary for the Birth Registry have been incorporated by some leading electronic medical record vendors and will eventually serve as the backbone of the Birth Registry. These clinical data elements can then be extracted directly from the electronic medical record without the need for manual abstraction by trained data collectors. Our goal in “Measuring Childbirth Outcomes Using Administrative and Birth Certificate Data” was to examine the feasibility of using lower-quality data (hospital administrative and birth certificate data) to create risk-adjusted outcome measures for childbirth outcomes while we await broad-based penetration of Birth Registry–compliant electronic medical records. Because hospitals collect administrative and birth certificate data on all births, such data could be used to provide expectant mothers with information on which to base their choice of providers, as well as providing clinicians with actionable performance feedback—while we await more robust clinical data based on the electronic medical record.